* [PATCH] x86: signal: check signal stack overflow properly
@ 2009-03-19 17:56 Hiroshi Shimamoto
2009-03-20 18:03 ` [tip:x86/signal] " Hiroshi Shimamoto
0 siblings, 1 reply; 6+ messages in thread
From: Hiroshi Shimamoto @ 2009-03-19 17:56 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin; +Cc: linux-kernel
From: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Impact: cleanup
Check alternate signal stack overflow with proper stack pointer. The stack
pointer of the next signal frame is different if that task has i387 state.
On x86_64, redzone would be included.
No need to check SA_ONSTACK if we're already using alternate signal stack.
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
---
arch/x86/kernel/signal.c | 48 +++++++++++++++++++++++++--------------------
1 files changed, 27 insertions(+), 21 deletions(-)
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index 4da412a..62f2164 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -211,31 +211,27 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
{
/* Default to using normal stack */
unsigned long sp = regs->sp;
+ int onsigstack = on_sig_stack(sp);
#ifdef CONFIG_X86_64
/* redzone */
sp -= 128;
#endif /* CONFIG_X86_64 */
- /*
- * If we are on the alternate signal stack and would overflow it, don't.
- * Return an always-bogus address instead so we will die with SIGSEGV.
- */
- if (on_sig_stack(sp) && !likely(on_sig_stack(sp - frame_size)))
- return (void __user *) -1L;
-
- /* This is the X/Open sanctioned signal stack switching. */
- if (ka->sa.sa_flags & SA_ONSTACK) {
- if (sas_ss_flags(sp) == 0)
- sp = current->sas_ss_sp + current->sas_ss_size;
- } else {
+ if (!onsigstack) {
+ /* This is the X/Open sanctioned signal stack switching. */
+ if (ka->sa.sa_flags & SA_ONSTACK) {
+ if (sas_ss_flags(sp) == 0)
+ sp = current->sas_ss_sp + current->sas_ss_size;
+ } else {
#ifdef CONFIG_X86_32
- /* This is the legacy signal stack switching. */
- if ((regs->ss & 0xffff) != __USER_DS &&
- !(ka->sa.sa_flags & SA_RESTORER) &&
- ka->sa.sa_restorer)
- sp = (unsigned long) ka->sa.sa_restorer;
+ /* This is the legacy signal stack switching. */
+ if ((regs->ss & 0xffff) != __USER_DS &&
+ !(ka->sa.sa_flags & SA_RESTORER) &&
+ ka->sa.sa_restorer)
+ sp = (unsigned long) ka->sa.sa_restorer;
#endif /* CONFIG_X86_32 */
+ }
}
if (used_math()) {
@@ -244,12 +240,22 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
sp = round_down(sp, 64);
#endif /* CONFIG_X86_64 */
*fpstate = (void __user *)sp;
-
- if (save_i387_xstate(*fpstate) < 0)
- return (void __user *)-1L;
}
- return (void __user *)align_sigframe(sp - frame_size);
+ sp = align_sigframe(sp - frame_size);
+
+ /*
+ * If we are on the alternate signal stack and would overflow it, don't.
+ * Return an always-bogus address instead so we will die with SIGSEGV.
+ */
+ if (onsigstack && !likely(on_sig_stack(sp)))
+ return (void __user *)-1L;
+
+ /* save i387 state */
+ if (used_math() && save_i387_xstate(*fpstate) < 0)
+ return (void __user *)-1L;
+
+ return (void __user *)sp;
}
#ifdef CONFIG_X86_32
--
1.6.1.2
^ permalink raw reply related [flat|nested] 6+ messages in thread* [tip:x86/signal] x86: signal: check signal stack overflow properly
2009-03-19 17:56 [PATCH] x86: signal: check signal stack overflow properly Hiroshi Shimamoto
@ 2009-03-20 18:03 ` Hiroshi Shimamoto
2009-03-24 22:00 ` Oleg Nesterov
2009-04-09 12:22 ` [tip:x86/signal] x86: signal: check signal stack overflow properly Jaswinder Singh Rajput
0 siblings, 2 replies; 6+ messages in thread
From: Hiroshi Shimamoto @ 2009-03-20 18:03 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, h-shimamoto, hpa, mingo, roland, tglx, mingo
Commit-ID: 14fc9fbc700dc95b4f46ebd588169324fe6deff8
Gitweb: http://git.kernel.org/tip/14fc9fbc700dc95b4f46ebd588169324fe6deff8
Author: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
AuthorDate: Thu, 19 Mar 2009 10:56:29 -0700
Committer: Ingo Molnar <mingo@elte.hu>
CommitDate: Fri, 20 Mar 2009 19:01:31 +0100
x86: signal: check signal stack overflow properly
Impact: cleanup
Check alternate signal stack overflow with proper stack pointer.
The stack pointer of the next signal frame is different if that
task has i387 state.
On x86_64, redzone would be included.
No need to check SA_ONSTACK if we're already using alternate signal stack.
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Cc: Roland McGrath <roland@redhat.com>
LKML-Reference: <49C2874D.3080002@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
arch/x86/kernel/signal.c | 48 +++++++++++++++++++++++++--------------------
1 files changed, 27 insertions(+), 21 deletions(-)
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index d2cc642..dfcc74a 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -211,31 +211,27 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
{
/* Default to using normal stack */
unsigned long sp = regs->sp;
+ int onsigstack = on_sig_stack(sp);
#ifdef CONFIG_X86_64
/* redzone */
sp -= 128;
#endif /* CONFIG_X86_64 */
- /*
- * If we are on the alternate signal stack and would overflow it, don't.
- * Return an always-bogus address instead so we will die with SIGSEGV.
- */
- if (on_sig_stack(sp) && !likely(on_sig_stack(sp - frame_size)))
- return (void __user *) -1L;
-
- /* This is the X/Open sanctioned signal stack switching. */
- if (ka->sa.sa_flags & SA_ONSTACK) {
- if (sas_ss_flags(sp) == 0)
- sp = current->sas_ss_sp + current->sas_ss_size;
- } else {
+ if (!onsigstack) {
+ /* This is the X/Open sanctioned signal stack switching. */
+ if (ka->sa.sa_flags & SA_ONSTACK) {
+ if (sas_ss_flags(sp) == 0)
+ sp = current->sas_ss_sp + current->sas_ss_size;
+ } else {
#ifdef CONFIG_X86_32
- /* This is the legacy signal stack switching. */
- if ((regs->ss & 0xffff) != __USER_DS &&
- !(ka->sa.sa_flags & SA_RESTORER) &&
- ka->sa.sa_restorer)
- sp = (unsigned long) ka->sa.sa_restorer;
+ /* This is the legacy signal stack switching. */
+ if ((regs->ss & 0xffff) != __USER_DS &&
+ !(ka->sa.sa_flags & SA_RESTORER) &&
+ ka->sa.sa_restorer)
+ sp = (unsigned long) ka->sa.sa_restorer;
#endif /* CONFIG_X86_32 */
+ }
}
if (used_math()) {
@@ -244,12 +240,22 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
sp = round_down(sp, 64);
#endif /* CONFIG_X86_64 */
*fpstate = (void __user *)sp;
-
- if (save_i387_xstate(*fpstate) < 0)
- return (void __user *)-1L;
}
- return (void __user *)align_sigframe(sp - frame_size);
+ sp = align_sigframe(sp - frame_size);
+
+ /*
+ * If we are on the alternate signal stack and would overflow it, don't.
+ * Return an always-bogus address instead so we will die with SIGSEGV.
+ */
+ if (onsigstack && !likely(on_sig_stack(sp)))
+ return (void __user *)-1L;
+
+ /* save i387 state */
+ if (used_math() && save_i387_xstate(*fpstate) < 0)
+ return (void __user *)-1L;
+
+ return (void __user *)sp;
}
#ifdef CONFIG_X86_32
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [tip:x86/signal] x86: signal: check signal stack overflow properly
2009-03-20 18:03 ` [tip:x86/signal] " Hiroshi Shimamoto
@ 2009-03-24 22:00 ` Oleg Nesterov
2009-03-26 17:03 ` Hiroshi Shimamoto
2009-04-09 12:22 ` [tip:x86/signal] x86: signal: check signal stack overflow properly Jaswinder Singh Rajput
1 sibling, 1 reply; 6+ messages in thread
From: Oleg Nesterov @ 2009-03-24 22:00 UTC (permalink / raw)
To: Hiroshi Shimamoto
Cc: linux-tip-commits, linux-kernel, hpa, mingo, roland, tglx, mingo
On 03/20, Hiroshi Shimamoto wrote:
>
> Commit-ID: 14fc9fbc700dc95b4f46ebd588169324fe6deff8
> Gitweb: http://git.kernel.org/tip/14fc9fbc700dc95b4f46ebd588169324fe6deff8
> Author: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
> AuthorDate: Thu, 19 Mar 2009 10:56:29 -0700
> Committer: Ingo Molnar <mingo@elte.hu>
> CommitDate: Fri, 20 Mar 2009 19:01:31 +0100
>
> x86: signal: check signal stack overflow properly
>
> Impact: cleanup
>
> Check alternate signal stack overflow with proper stack pointer.
> The stack pointer of the next signal frame is different if that
> task has i387 state.
I think the patch is correct but I have a minor question,
> No need to check SA_ONSTACK if we're already using alternate signal stack.
Yes, but this also mean that we don't need sas_ss_flags() under
"if (!onsigstack)",
> @@ -211,31 +211,27 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
> {
> /* Default to using normal stack */
> unsigned long sp = regs->sp;
> + int onsigstack = on_sig_stack(sp);
>
> #ifdef CONFIG_X86_64
> /* redzone */
> sp -= 128;
> #endif /* CONFIG_X86_64 */
>
> - /*
> - * If we are on the alternate signal stack and would overflow it, don't.
> - * Return an always-bogus address instead so we will die with SIGSEGV.
> - */
> - if (on_sig_stack(sp) && !likely(on_sig_stack(sp - frame_size)))
> - return (void __user *) -1L;
> -
> - /* This is the X/Open sanctioned signal stack switching. */
> - if (ka->sa.sa_flags & SA_ONSTACK) {
> - if (sas_ss_flags(sp) == 0)
> - sp = current->sas_ss_sp + current->sas_ss_size;
> - } else {
> + if (!onsigstack) {
> + /* This is the X/Open sanctioned signal stack switching. */
> + if (ka->sa.sa_flags & SA_ONSTACK) {
> + if (sas_ss_flags(sp) == 0)
> + sp = current->sas_ss_sp + current->sas_ss_size;
We can use "->sas_ss_size != 0" instead and avoid the unnecessary
sas_ss_flags()->on_sig_stack() check.
Please note that afaics sas_ss_flags()->on_sig_stack() is actually
wrong because we already adjusted "sp" above for redzone.
Suppose that on_sig_stack(regs->sp) = F, but "sp - 128" falls into
the altstack. In that case SA_ONSTACK won't switch the stack.
Of course, this is only theoretical, but still.
Oleg.
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [tip:x86/signal] x86: signal: check signal stack overflow properly
2009-03-24 22:00 ` Oleg Nesterov
@ 2009-03-26 17:03 ` Hiroshi Shimamoto
2009-04-01 15:16 ` [tip:x86/urgent] x86: signal: check sas_ss_size instead of sas_ss_flags() Hiroshi Shimamoto
0 siblings, 1 reply; 6+ messages in thread
From: Hiroshi Shimamoto @ 2009-03-26 17:03 UTC (permalink / raw)
To: Oleg Nesterov
Cc: linux-tip-commits, linux-kernel, hpa, mingo, roland, tglx, mingo
Oleg Nesterov wrote:
> On 03/20, Hiroshi Shimamoto wrote:
>> Commit-ID: 14fc9fbc700dc95b4f46ebd588169324fe6deff8
>> Gitweb: http://git.kernel.org/tip/14fc9fbc700dc95b4f46ebd588169324fe6deff8
>> Author: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
>> AuthorDate: Thu, 19 Mar 2009 10:56:29 -0700
>> Committer: Ingo Molnar <mingo@elte.hu>
>> CommitDate: Fri, 20 Mar 2009 19:01:31 +0100
>>
>> x86: signal: check signal stack overflow properly
>>
>> Impact: cleanup
>>
>> Check alternate signal stack overflow with proper stack pointer.
>> The stack pointer of the next signal frame is different if that
>> task has i387 state.
>
> I think the patch is correct but I have a minor question,
>
>> No need to check SA_ONSTACK if we're already using alternate signal stack.
>
> Yes, but this also mean that we don't need sas_ss_flags() under
> "if (!onsigstack)",
>
>> @@ -211,31 +211,27 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
>> {
>> /* Default to using normal stack */
>> unsigned long sp = regs->sp;
>> + int onsigstack = on_sig_stack(sp);
>>
>> #ifdef CONFIG_X86_64
>> /* redzone */
>> sp -= 128;
>> #endif /* CONFIG_X86_64 */
>>
>> - /*
>> - * If we are on the alternate signal stack and would overflow it, don't.
>> - * Return an always-bogus address instead so we will die with SIGSEGV.
>> - */
>> - if (on_sig_stack(sp) && !likely(on_sig_stack(sp - frame_size)))
>> - return (void __user *) -1L;
>> -
>> - /* This is the X/Open sanctioned signal stack switching. */
>> - if (ka->sa.sa_flags & SA_ONSTACK) {
>> - if (sas_ss_flags(sp) == 0)
>> - sp = current->sas_ss_sp + current->sas_ss_size;
>> - } else {
>> + if (!onsigstack) {
>> + /* This is the X/Open sanctioned signal stack switching. */
>> + if (ka->sa.sa_flags & SA_ONSTACK) {
>> + if (sas_ss_flags(sp) == 0)
>> + sp = current->sas_ss_sp + current->sas_ss_size;
>
> We can use "->sas_ss_size != 0" instead and avoid the unnecessary
> sas_ss_flags()->on_sig_stack() check.
>
> Please note that afaics sas_ss_flags()->on_sig_stack() is actually
> wrong because we already adjusted "sp" above for redzone.
>
> Suppose that on_sig_stack(regs->sp) = F, but "sp - 128" falls into
> the altstack. In that case SA_ONSTACK won't switch the stack.
>
> Of course, this is only theoretical, but still.
Hi Oleg,
Thanks for pointing out it.
I made a patch you suggested.
I haven't tested enough this patch, sorry.
Thanks,
Hiroshi
========
From: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Subject: [PATCH] x86: signal: check sas_ss_size instead of sas_ss_flags()
Impact: fix redundant and incorrect check
Checking on_sig_stack() in sas_ss_flags() at get_sigframe() is redundant
and not correct on 64 bit. To check sas_ss_size is enough.
Reported-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
---
arch/x86/kernel/signal.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index 62f2164..465b42d 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -221,7 +221,7 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
if (!onsigstack) {
/* This is the X/Open sanctioned signal stack switching. */
if (ka->sa.sa_flags & SA_ONSTACK) {
- if (sas_ss_flags(sp) == 0)
+ if (current->sas_ss_size)
sp = current->sas_ss_sp + current->sas_ss_size;
} else {
#ifdef CONFIG_X86_32
--
1.6.1.2
^ permalink raw reply related [flat|nested] 6+ messages in thread* [tip:x86/urgent] x86: signal: check sas_ss_size instead of sas_ss_flags()
2009-03-26 17:03 ` Hiroshi Shimamoto
@ 2009-04-01 15:16 ` Hiroshi Shimamoto
0 siblings, 0 replies; 6+ messages in thread
From: Hiroshi Shimamoto @ 2009-04-01 15:16 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, h-shimamoto, hpa, mingo, oleg, tglx, mingo
Commit-ID: 0f8f308925ebe0480bd9831d32963ee0b885e24b
Gitweb: http://git.kernel.org/tip/0f8f308925ebe0480bd9831d32963ee0b885e24b
Author: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
AuthorDate: Thu, 26 Mar 2009 10:03:08 -0700
Committer: Ingo Molnar <mingo@elte.hu>
CommitDate: Wed, 1 Apr 2009 17:13:17 +0200
x86: signal: check sas_ss_size instead of sas_ss_flags()
Impact: fix redundant and incorrect check
Oleg Nesterov noticed wrt commit:
14fc9fb: x86: signal: check signal stack overflow properly
>> No need to check SA_ONSTACK if we're already using alternate signal stack.
>
> Yes, but this also mean that we don't need sas_ss_flags() under
> "if (!onsigstack)",
Checking on_sig_stack() in sas_ss_flags() at get_sigframe() is redundant
and not correct on 64 bit. To check sas_ss_size is enough.
Reported-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Cc: roland@redhat.com
LKML-Reference: <49CBB54C.5080201@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
arch/x86/kernel/signal.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index dfcc74a..1442516 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -221,7 +221,7 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
if (!onsigstack) {
/* This is the X/Open sanctioned signal stack switching. */
if (ka->sa.sa_flags & SA_ONSTACK) {
- if (sas_ss_flags(sp) == 0)
+ if (current->sas_ss_size)
sp = current->sas_ss_sp + current->sas_ss_size;
} else {
#ifdef CONFIG_X86_32
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [tip:x86/signal] x86: signal: check signal stack overflow properly
2009-03-20 18:03 ` [tip:x86/signal] " Hiroshi Shimamoto
2009-03-24 22:00 ` Oleg Nesterov
@ 2009-04-09 12:22 ` Jaswinder Singh Rajput
1 sibling, 0 replies; 6+ messages in thread
From: Jaswinder Singh Rajput @ 2009-04-09 12:22 UTC (permalink / raw)
To: mingo, hpa, h-shimamoto, linux-kernel, roland, tglx, mingo
Cc: linux-tip-commits
Hiroshi-san,
I am getting few sparse warnings from your commited lines.
If possible, please run sparse before sending patch.
On Fri, 2009-03-20 at 18:03 +0000, Hiroshi Shimamoto wrote:
> Commit-ID: 14fc9fbc700dc95b4f46ebd588169324fe6deff8
> Gitweb: http://git.kernel.org/tip/14fc9fbc700dc95b4f46ebd588169324fe6deff8
> Author: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
> AuthorDate: Thu, 19 Mar 2009 10:56:29 -0700
> Committer: Ingo Molnar <mingo@elte.hu>
> CommitDate: Fri, 20 Mar 2009 19:01:31 +0100
>
> x86: signal: check signal stack overflow properly
>
> Impact: cleanup
>
> Check alternate signal stack overflow with proper stack pointer.
> The stack pointer of the next signal frame is different if that
> task has i387 state.
>
> On x86_64, redzone would be included.
>
> No need to check SA_ONSTACK if we're already using alternate signal stack.
>
> Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
> Cc: Roland McGrath <roland@redhat.com>
> LKML-Reference: <49C2874D.3080002@ct.jp.nec.com>
> Signed-off-by: Ingo Molnar <mingo@elte.hu>
>
>
> ---
> arch/x86/kernel/signal.c | 48 +++++++++++++++++++++++++--------------------
> 1 files changed, 27 insertions(+), 21 deletions(-)
>
> diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
> index d2cc642..dfcc74a 100644
> --- a/arch/x86/kernel/signal.c
> +++ b/arch/x86/kernel/signal.c
> @@ -244,12 +240,22 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
> sp = round_down(sp, 64);
> #endif /* CONFIG_X86_64 */
> *fpstate = (void __user *)sp;
> -
> - if (save_i387_xstate(*fpstate) < 0)
> - return (void __user *)-1L;
> }
>
> - return (void __user *)align_sigframe(sp - frame_size);
> + sp = align_sigframe(sp - frame_size);
> +
> + /*
> + * If we are on the alternate signal stack and would overflow it, don't.
> + * Return an always-bogus address instead so we will die with SIGSEGV.
> + */
> + if (onsigstack && !likely(on_sig_stack(sp)))
> + return (void __user *)-1L;
arch/x86/kernel/signal.c:251:11: warning: cast adds address space to expression (<asn:1>)
> +
> + /* save i387 state */
> + if (used_math() && save_i387_xstate(*fpstate) < 0)
> + return (void __user *)-1L;
arch/x86/kernel/signal.c:255:11: warning: cast adds address space to expression (<asn:1>)
> +
> + return (void __user *)sp;
> }
>
> #ifdef CONFIG_X86_32
4a612048 arch/x86/kernel/signal_32.c (Hiroshi Shimamoto 2008-11-11 19:09:29 -0800 326) err |= __put_user(*((u64 *)&retcode), (u64 *)frame->retcode);
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: incorrect type in argument 1 (different address spaces)
arch/x86/kernel/signal.c:326:9: expected void const volatile [noderef] <asn:1>*<noident>
arch/x86/kernel/signal.c:326:9: got unsigned long long [usertype] *<noident>
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast adds address space to expression (<asn:1>)
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast adds address space to expression (<asn:1>)
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast adds address space to expression (<asn:1>)
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
arch/x86/kernel/signal.c:326:9: warning: cast removes address space of expression
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 359) put_user_try {
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 360) put_user_ex(sig, &frame->sig);
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 361) put_user_ex(&frame->info, &frame->pinfo);
arch/x86/kernel/signal.c:361:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:361:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:361:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:361:3: warning: cast removes address space of expression
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 362) put_user_ex(&frame->uc, &frame->puc);
arch/x86/kernel/signal.c:362:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:362:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:362:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:362:3: warning: cast removes address space of expression
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 363) err |= copy_siginfo_to_user(&frame->info, info);
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 364)
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 379) /* Set up to return from userspace. */
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 380) restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_si
arch/x86/kernel/signal.c:380:12: warning: incorrect type in assignment (different address spaces)
arch/x86/kernel/signal.c:380:12: expected void [noderef] <asn:1>*restorer
arch/x86/kernel/signal.c:380:12: got void *<noident>
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 381) if (ka->sa.sa_flags & SA_RESTORER)
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 382) restorer = ka->sa.sa_restorer;
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 383) put_user_ex(restorer, &frame->pretcode);
arch/x86/kernel/signal.c:383:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:383:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:383:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:383:3: warning: cast removes address space of expression
98e3d45e arch/x86/kernel/signal.c (Hiroshi Shimamoto 2009-01-23 15:50:10 -0800 392) put_user_ex(*((u64 *)&rt_retcode), (u64 *)frame->retcode)
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: incorrect type in argument 1 (different address spaces)
arch/x86/kernel/signal.c:392:3: expected void const volatile [noderef] <asn:1>*<noident>
arch/x86/kernel/signal.c:392:3: got unsigned long long [usertype] *<noident>
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast adds address space to expression (<asn:1>)
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast adds address space to expression (<asn:1>)
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast adds address space to expression (<asn:1>)
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
arch/x86/kernel/signal.c:392:3: warning: cast removes address space of expression
--
JSR
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2009-04-09 12:22 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-03-19 17:56 [PATCH] x86: signal: check signal stack overflow properly Hiroshi Shimamoto
2009-03-20 18:03 ` [tip:x86/signal] " Hiroshi Shimamoto
2009-03-24 22:00 ` Oleg Nesterov
2009-03-26 17:03 ` Hiroshi Shimamoto
2009-04-01 15:16 ` [tip:x86/urgent] x86: signal: check sas_ss_size instead of sas_ss_flags() Hiroshi Shimamoto
2009-04-09 12:22 ` [tip:x86/signal] x86: signal: check signal stack overflow properly Jaswinder Singh Rajput
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.